Understanding Emojis for Financial Sentiment Analysis

2023 | International Conference on Information Systems | Citations: 0

Authors: Chen, Siyi; Xing, Frank

Abstract: ... Expand

Abstract: Collapse

Topics: application programming interface social media website cryptocurrency Python

Methods: sentiment analysis machine learning BERT deep learning experiment

Text Analytics

2023 | HICSS | Citations: 0

Authors: Cogburn, D.; Hine, M.; Yoon, V.

Abstract: This virtual minitrack for HICSS-56 recognizes that most global collaboration s ... Expand

Abstract: This virtual minitrack for HICSS-56 recognizes that most global collaboration systems, social media, and information systems of all types, generate enormous amounts of textual data, including system logs, email archives, websites, blog posts, meeting transcripts, speeches, annual reports, published material, and social media posts.While this structured, semistructured, and unstructured textual data is readily available, it presents tremendous challenges to researchers trying to analyze these large bodies of text with traditional methods. Text mining in big data analytics is an increasingly important technique for an interdisciplinary group of scholars, practitioners, government officials, and international organizations. This minitrack explores the tools, techniques, and insights generated from the analysis of text data. Collapse

Topics: application programming interface

Methods: natural language processing RoBERTa experiment question answering classification method

Incorporating Multiple Knowledge Sources for Targeted Aspect-based Financial Sentiment Analysis

2023 | ACM Transactions on Management Information Systems | Citations: 0

Authors: Du, Kelvin; Xing, Frank; Cambria, Erik

Abstract: Combining symbolic and subsymbolic methods has become a promising strategy as re ... Expand

Abstract: Combining symbolic and subsymbolic methods has become a promising strategy as research tasks in AI grow increasingly complicated and require higher levels of understanding. Targeted Aspect-based Financial Sentiment Analysis (TABFSA) is an example of such complicated tasks, as it involves processes like information extraction, information specification, and domain adaptation. However, little is known about the design principles of such hybrid models leveraging external lexical knowledge. To fill this gap, we define anterior, parallel, and posterior knowledge integration and propose incorporating multiple lexical knowledge sources strategically into the fine-tuning process of pre-trained transformer models for TABFSA. Experiments on the Financial Opinion mining and Question Answering challenge (FiQA) Task 1 and SemEval 2017 Task 5 datasets show that the knowledge-enabled models systematically improve upon their plain deep learning counterparts, and some outperform state-of-the-art results reported in terms of aspect sentiment analysis error. We discover that parallel knowledge integration is the most effective and domain-specific lexical knowledge is more important according to our ablation analysis. Collapse

Topics: microblog domain-specific language Twitter social media knowledge graph

Methods: sentiment analysis RoBERTa transformer model BERT XLNet

Building a deep learning-based QA system from a CQA dataset

2023 | Decision Support Systems | Citations: 0

Authors: Jin, Sol; Lian, Xu; Jung, Hanearl; Park, Jinsoo; Suh, Jihae

Abstract: A man-made machine-reading comprehension (MRC) dataset is necessary to train the ... Expand

Abstract: A man-made machine-reading comprehension (MRC) dataset is necessary to train the answer extraction part of existing Question Answering (QA) systems. However, a high-quality and well-structured dataset with questionparagraph-answer pairs is not usually found in the real world. Furthermore, updating or building an MRC dataset is a challenging and costly affair. To address these shortcomings, we propose a QA system that uses a large-scale English Community Question Answering (CQA) dataset (i.e., Stack Exchange) composed of 3,081,834 questionanswer pairs. The QA system adopts a classifier-retriever-summarizer structure design. The question classifier and the answer retriever part are based on a Bidirectional Encoder Representations from Transformers (BERT) Natural Language Processing (NLP) model by Google, and the summarizer part introduces a deep learning-based Text-to-Text Transfer Transformer (T5) model to summarize the long answers. We instantiated the proposed QA system with 140 topics from the CQA dataset (including topics such as biology, law, politics, etc.) and conducted human and automatic evaluations. Our system presented encouraging results, considering that it provides highquality answers to the questions in the test set and satisfied the requirements to develop a QA system without MRC datasets. Our results show the potential of building automatic and high-performance QA systems without being limited by man-made datasets, a significant step forward in the research of open-domain or specificdomain QA systems. Collapse

Topics: privacy decision making online community IT productivity Google+

Methods: BERT chatGPT computational algorithm experiment natural language processing

Scalable Sustainability Monitoring of Financial Reports: A Design Science Artifact

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Rath, Oliver; Haase, Frederic; Erlacher, Nina Annika; Schoder, Detlef

Abstract: The United Nations Climate Change (UNCC) report in 2022 warns that the likeliho ... Expand

Abstract: The United Nations Climate Change (UNCC) report in 2022 warns that the likelihood of global warming exceeding 1.5°C by 2026 has surpassed 50%. Climate change is a global problem that requires collective action, and companies have a significant role to play in mitigating climate change. Companies have already started to incorporate climate change risk and liabilities in their annual reports. Monitoring a company's progress in transitioning to sustainability can help stakeholders make informed decisions. However, existing approaches for sustainability monitoring of annual reports are limited in scalability, mainly because of the manual steps involved. Therefore, we designed and evaluated a more scalable artifact using state-of-the-art natural language processing (NLP) techniques to monitor companies' sustainability targets based on their annual reports. This work presents a prototype that improves upon the current state of practice and contributes to the body of knowledge by outlining key design decisions. Collapse

Topics: usability evaluation criteria information portal computer data storage artificial intelligence

Methods: natural language processing computational algorithm question answering design artifact machine learning

May the Guide Be With You: CA-facilitated Information Elicitation to Prevent Service Failure

2023 | European Conference On Information Systems | Citations: 0

Authors: Poser, Mathis; Hörhold, Hauke; Bittner, Eva

Abstract: Companies automate the delivery of their online services by deploying artificia ... Expand

Abstract: Companies automate the delivery of their online services by deploying artificial intelligence-based conversational agents (CAs). However, contemporary CAs still struggle to reliably answer the full range of requests from support seekers. To avoid service failure, service delivery activities of CAs and service employees should be interconnected by a handover of requests. This form of hybrid service delivery requires support seekers to disclose relevant information so that CAs can relay them to service employees prior to an imminent failure. By integrating and extending design knowledge from two DSR projects, we derive four design principles (DPs) to prepare handovers. These DPs guided the implementation of a service script in a CA prototype to facilitate the elicitation of information from support seekers. Based on two evaluation episodes, we show that support seekers feel supported by the CA in disclosing information which results in elaborate input for subsequent processing by service employees after handover. Collapse

Topics: system support artificial intelligence user satisfaction website data quality

Methods: design artifact usability testing survey qualitative content analysis design science

Navy Messaging Analytics to Improve Readiness, Response and Preparedness

2022 | HICSS | Citations: 0

Authors: Dilmaghani, Raheleh; Turco, Kyle; Diego, San

Abstract: There are tens of thousands of official messages exchanged within the Navy’s mess ... Expand

Abstract: There are tens of thousands of official messages exchanged within the Navy’s messaging system daily. The messages carry a tremendous amount of data that is not fully processed nor analyzed. If we could enhance extracting the value of these data, and automate, we could improve speed, readiness, performance, maintenance, outage scheduling, and much more. Although messages follow a prescribed format, it is not uniform across the message types and data entry is not enforced and prone to human interpretation or error in containing all relevant information and correct categorization or labeling. Often there is relevant information in other messages not directly linked or referenced, but if inferred could enhance the observer’s perspective and response to the situation. By analyzing the massive data stored within Navy tactical messages, we could learn about fleet readiness without asking a particular unit or group, “are you ready?” Better yet, we may be able to predict future readiness based on past schedule and performance data if correlated correctly to learn patterns that indicate future mission success. In this paper, we present results from a pilot project where we show the significance of processing available messages for relevant information by not limiting the message types that seem to be the right category or to those with an explicit keyword. This process helps to detect relevant information and create meaning correlations, which could contribute to more informed decision making or to provide the Commander information to make decisions proactively instead of reacting to negative outcomes like casualties or mishaps. Extracting relevant information in turn can contribute to more informed decision making in saving lives, reduced sustainment costs, or repair costs. Collapse

Topics: electronic mail Python analytical information system knowledge base automated reasoning

Methods: support vector machine machine learning logistic regression supervised learning natural language processing

Where was COVID-19 first discovered? Designing a question-answering system for pandemic situations

2022 | European Conference On Information Systems | Citations: 0

Authors: Graf, Johannes; Lancho, Gino; Zschech, Patrick; Heinrich, Kai

Abstract: The COVID-19 pandemic is accompanied by a massive "infodemic" that makes it har ... Expand

Abstract: The COVID-19 pandemic is accompanied by a massive "infodemic" that makes it hard to identify concise and credible information for COVID-19-related questions, like incubation time, infection rates, or the effectiveness of vaccines. As a novel solution, our paper is concerned with designing a question-answering system based on modern technologies from natural language processing to overcome information overload and misinformation in pandemic situations. To carry out our research, we followed a design science research approach and applied Ingwersen's cognitive model of information retrieval interaction to inform our design process from a socio-technical lens. On this basis, we derived prescriptive design knowledge in terms of design requirements and design principles, which we translated into the construction of a prototypical instantiation. Our implementation is based on the comprehensive CORD-19 dataset, and we demonstrate our artifact's usefulness by evaluating its answer quality based on a sample of COVID-19 questions labeled by biomedical experts. Collapse

Topics: knowledge base pandemic covid 19 database system Elasticsearch

Methods: design artifact design principle design science question answering design requirement

Supporting Online Customer Feedback Management with Automatic Review Response Generation

2022 | HICSS | Citations: 1

Authors: Katsiuba, Dzmitry; Kew, Tannon; Dolata, Mateusz; Schwabe, Gerhard

Abstract: The growing amount of online reviews plays a significant role in a business' ima ... Expand

Abstract: The growing amount of online reviews plays a significant role in a business' image and performance. Businesses in the hospitality industry often lack necessary resources to organize and manage online customer feedback and are therefore likely to search for alternative ways to handle this. AI-based technologies may offer valuable solutions. However, there is currently little research on if and how AI solutions may support the process of responding to online customer feedback in the hospitality industry. This paper presents and evaluates a concept for assisting customer feedback management with automatically generated responses to online reviews. Our solution contributes to ongoing investigations into text generation applications for supporting human authors and also proposes new approaches and potential business models for managing online customer feedback. Collapse

Topics: artificial intelligence online review business model outsourcing Turing test

Methods: survey natural language processing survey design descriptive statistic design artifact

Designing for Conversational System Trustworthiness: The Impact of Model Transparency on Trust and Task Performance

2022 | European Conference On Information Systems | Citations: 1

Authors: Schmitt, Anuschka; Wambsganss, Thiemo; Janson, Andreas

Abstract: Designing for system trustworthiness promises to address challenges of opaquene ... Expand

Abstract: Designing for system trustworthiness promises to address challenges of opaqueness and uncertainty introduced through Machine Learning (ML)-based systems by allowing users to understand and interpret systems' underlying working mechanisms. However, empirical exploration of trustworthiness measures and their effectiveness is scarce and inconclusive. We investigated how varying model confidence (70% versus 90%) and making confidence levels transparent to the user (explanatory statement versus no explanatory statement) may influence perceptions of trust and performance in an information retrieval task assisted by a conversational system. In a field experiment with 104 users, our findings indicate that neither model confidence nor transparency seem to impact trust in the conversational system. However, users' task performance is positively influenced by both transparency and trust in the system. While this study considers the complex interplay of system trustworthiness, trust, and subsequent behavioral outcomes, our results call into question the relation between system trustworthiness and user trust. Collapse

Topics: task productivity information retrieval usability human computer interaction advertising management

Methods: experiment survey ANOVA parametric test chatbot

Theories: trust theory big five model